The overall objectives of this research are to develop methods (1) for estimating vaccine efficacy and effectiveness in the field and (2) for characterizing complex and long-term properties of vaccination in individuals and populations.
The specific aims are 1. To continue development of study designs and methods of analysis for using validation samples to correct for misclassified or partially observed outcomes and missing data in vaccine studies. The goal is improved estimates of protective vaccine efficacy, VES, and vaccine efficacy for infectiousness, VEI, as well as designs for efficient, cost-effective studies. 2. To continue development of cluster-randomized designs to evaluate indirect, total, and overall effectiveness of vaccination strategies. This includes methods for optimal spatial randomization constraints. A new stepped wedge study using longitudinal and cross-sectional carriage studies as baseline data will be developed. The new mini-community design to evaluate indirect effects will be developed in which the cluster is a household or other small transmission unit. 3. To develop methods for joint estimation of the protective effects of vaccination, VES and vaccine efficacy for infectiousness, VEI from households embedded within clusters in a cluster-randomized trial. 4. To develop methods to evaluate immunological correlates of protection that combine models accounting for exposure to infection with approaches delineating different degrees of confidence about the correlate. 5. To continue to explore interpretation of the protective effects of vaccination, VES, when combining results across studies in different populations, taking into account different levels of baseline transmission and pre-existing immunity. This includes developing methods for evaluating vaccine efficacy when a vaccine puts evolutionary pressure on the composition of the population of circulating strains. Statistical approaches to be developed include likelihood inference, Bayesian methods, semiparametric methods, hierarchical models, and survival methods. These methods will be motivated by the design and analysis of studies of influenza, pneumococcal, meningococcal A, rotavirus, and malaria transmission blocking vaccines.
The overall objectives of this research are to develop statistical and epidemiologic methods (1) for estimating vaccine efficacy and effectiveness in the field and (2) for characterizing complex and long-term properties of vaccination in individuals and populations. Our research is motivated by influenza, meningococcal A, pneumococcal, rotavirus, and malaria vaccine field studies on several continents.
|Gabriel, Erin E; Gilbert, Peter B (2014) Evaluating principal surrogate endpoints with time-to-event data accounting for time-varying treatment efficacy. Biostatistics 15:251-65|
|Halloran, M Elizabeth; Longini Jr, Ira M (2014) Emerging, evolving, and established infectious diseases and interventions. Science 345:1292-4|
|Yang, Yang; Halloran, M Elizabeth; Chen, Yanjun et al. (2014) A pathway EM-algorithm for estimating vaccine efficacy with a non-monotone validation set. Biometrics 70:568-78|
|Matrajt, Laura; Halloran, M Elizabeth; Longini Jr, Ira M (2013) Optimal vaccine allocation for the early mitigation of pandemic influenza. PLoS Comput Biol 9:e1002964|
|Auranen, Kari; Rinta-Kokko, Hanna; Halloran, M Elizabeth (2013) Estimating strain-specific and overall efficacy of polyvalent vaccines against recurrent pathogens from a cross-sectional study. Biometrics 69:235-44|
|Halloran, M Elizabeth; Hudgens, Michael G (2012) Causal inference for vaccine effects on infectiousness. Int J Biostat 8:|